With over 400,000 patients undergoing spine fusion surgery each year, many of whom have low bone density because of their advanced age, it is now clear that failure of the bone after a fusion procedure is a serious and increasingly common problem. This negative outcome of the primary fusion surgery requires the patient have a second surgery, and arises because the patient's vertebral bone is too weak to sustain the loads applied to it after the primary fusion surgery. The bone can fail in different ways, the most common modes of bone failure being: 1) fracture of the adjacent-level vertebra; 2) loosening of a pedicle screw due to local failure of the supporting bone; and 3) subsidence or fracture of the vertebral endplates. These failure modes have been difficult to avoid because, although spine surgeons have a variety of both surgical and non-surgical approaches to reduce the risk of bone failure, these risk-mitigating measures add complexity and expense to the surgery and surgeons do not have an objective test to justify their use or identify which measures would be most suitable for a given high-risk patient. Thus, in the larger context of minimizing the incidence of bone failure after a spine fusion surgery, the long-term goal of this overall SBIR project is to develop new clinical tests that: 1) pre-operatively identify patients at high-risk of bone failure secondary to a fusion surgery; and 2) assess risk for each potential mode of bone failure (fracture of the adjacent-level vertebra; loosening of a pedicle screw; subsidence or fracture of the vertebral endplates). Our overall approach is based on providing a detailed biomechanical analysis of the patient's vertebrae, performed as an add-on analysis of the patient's pre-operative diagnostic computed tomography (CT) scan of their spine. For this Phase-I portion of the project, Aim 1 will develop a set of virtual stress tests of the patient's vertebrae, using fnite element analysis of the patient's CT scan. These biomechanical tests will probe each of the different types of failure modes and will account also for patient-specific measures of spine curvature and loading as well as disc morphology. These multiple tests will then be used in Aim 2, in which we will perform a retrospective blinded case-control study on over 300 spine fusion patients, for whom pre-operative CT scans have previously been acquired and clinical outcomes are on file. Finally, in Aim 3, we will perform statistical analyses to identify which of the test outcomes are most predictive of the clinical outcomes, accounting also for various clinical risk factors. We also seek to test the hypothesis that prediction of clinical outcomes by our biomechanics-based test is better than by the current best practices, setting the stage for further development of the technology in a subsequent Phase-II project. Bone failure secondary to spine fusion surgery is a serious problem. If successful, these new tests would be cost-effective to implement and would have substantial positive impact on patient care.